Dissertationen zum Thema „MOBILedit“

I do not know the number of university credits, but entered it as 30 ECTS. This was an exchange student and I do not know if they were actually registered at KTH.

The convergence of future networks relies on the evolution of technology that enables seamless roaming abilities across non-heterogenous networks for mobile clients. This thesis presents an experimental study of a GPRS-WLAN integration scenario where the objective is to analyze various aspects of the issues related to charging, mobility, roaming and security between GPRS and WLAN networks. The mainly discussed integration scenario in this thesis is loosely coupled systems working on RADIUS platforms, which together with MobileIP and IPSec provides the mobile client with a secure and access-technology independent network access platform.

In order to accommodate GPRS client authentication for WLAN operators, there is a prominent need for the incorporation of necessary GPRS functionality into present AAA servers. RADIUS has been studied as the initial target for the implementation of a GPRS interface towards SMS-Cs and HLRs.The authentication of a mobile client is performed against a HLR/AuC in a GPRS network, either over SS7 links or through the incorporation of SIGTRAN protocols over SCTP. SIGTRANsolutions has the ability to join WLAN networks in a SS7 resource sharing model where the SS7 authentication signalling traffic is transported over IP networks to a Signalling Gateway acting as the logical interface against SS7 networks.

GPRS-WLAN accounting may be solved through direct roaming agreements between mobile operators and in such a case transport CDR’s over FTP between their billing systems. If roaming agreements does not exist, it may be viable to establish relationships between WLANs and brokers as well as mobile operators and brokers. The brokering model provides a scalable model that allows easier exchange of charging and billing information on an infrastructure based on WLAN and GPRS billing systems. Standardised transmission protocols for accounting information such as GTP’/TAP3 may be utilised in order to provide a generic billing exchange format between billing systems and operators.

Furthermore, different network architectures may have different requirements in order to accommodate GPRS clients with WLAN access. A few network architectures has been analysed, and the developed GPRS AAA Interface Daemon (GAID) has been put into context in order to present a generic GPRS-WLAN systems integration solution for WLAN operators.

The analysed solutions in this thesis give various possibilities for WLAN operators to setup wireless services for bypassing mobile clients. The implementational work provides a RADIUS platform, which can be enhanced with functionality that enables communication over any interface in the future.

Designing self-regulating machines that can see and comprehend various real world objects around it are the main purpose of the AI domain. Recently, there has been marked advancements in the field of deep learning to create state-of-the-art DNNs for various CV applications. It is challenging to deploy these DNNs into resource-constrained micro-controller units as often they are quite memory intensive. Design Space Exploration is a technique which makes CNN/DNN memory efficient and more flexible to be deployed into resource-constrained hardware. MobileNet is small DNN architecture which was designed for embedded and mobile vision, but still researchers faced many challenges in deploying this model into resource limited real-time processors.

This thesis, proposes three new DNN architectures, which are developed using the Design Space Exploration technique. The state-of-the art MobileNet baseline architecture is used as foundation to propose these DNN architectures in this study. They are enhanced versions of the baseline MobileNet architecture. DSE techniques like data augmentation, architecture tuning, and architecture modification have been done to improve the baseline architecture. First, the Thin MobileNet architecture is proposed which uses more intricate block modules as compared to the baseline MobileNet. It is a compact, efficient and flexible architecture with good model accuracy. To get a more compact models, the KilobyteNet and the Ultra-thin MobileNet DNN architecture is proposed. Interesting techniques like channel depth alteration and hyperparameter tuning are introduced along-with some of the techniques used for designing the Thin MobileNet. All the models are trained and validated from scratch on the CIFAR-10 dataset. The experimental results (training and testing) can be visualized using the live accuracy and logloss graphs provided by the Liveloss package. The Ultra-thin MobileNet model is more balanced in terms of the model accuracy and model size out of the three and hence it is deployed into the NXP i.MX RT1060 embedded hardware unit for image classification application.

Computer Vision is a mathematical tool formulated to extend human vision to machines. This tool can perform various tasks such as object classification, object tracking, motion estimation, and image segmentation. These tasks find their use in many applications, namely robotics, self-driving cars, augmented reality, and mobile applications. However, opposed to the traditional technique of incorporating handcrafted features to understand images, convolution neural networks are being used to perform the same function. Computer vision applications widely use CNNs due to their stellar performance in interpreting images. Over the years, there have been numerous advancements in machine learning, particularly to CNNs. However, the need to improve their accuracy, model size and complexity increased, making their deployment in restricted environments a challenge. Many researchers proposed techniques to reduce the size of CNN while still retaining its accuracy. Few of these include network quantization, pruning, low rank, and sparse decomposition and knowledge distillation. Some methods developed efficient models from scratch. This thesis achieves a similar goal using design space exploration techniques on the latest variant of MobileNets, MobileNet V3. Using Depthwise Pointwise Depthwise (DPD) blocks, escalation in the number of expansion filters in some layers and mish activation function MobileNet V3 is reduced to 84.96% in size and made 0.2% more accurate. Furthermore, it is deployed in NXP i.MX RT1060 for image classification on CIFAR-10 dataset.

The SPECT-CT machines manufactured by Siemens consists of two heavy detector heads(~1500lbs each) that are moved into various configurations for radionuclide imaging. These detectors are driven by large torque powered by motors in the gantry that enable linear and rotational motion. If the detectors collide with large objects – stools, tables, patient extremities, etc. – they are very likely to damage the objects and get damaged as well. This research work proposes an intelligent real-time object detection system to prevent collisions between detector heads and external objects in the path of the detector’s motion by implementing an end-to-end deep learning object detector. The research extensively documents all the work done in identifying the most suitable object detection framework for this use case, collecting, and processing the image dataset of target objects, training the deep neural net to detect target objects, deploying the trained deep neural net in live demos by implementing a real-time object detection application written in Python, improving the model’s performance, and finally investigating methods to stop detector motion upon detecting external objects in the collision region. We successfully demonstrated that a Caffe version of MobileNet-SSD can be trained and deployed to detect target objects entering the collision region in real-time by following the methodologies outlined in this paper. We then laid out the future work that must be done in order to bring this system into production, such as training the model to detect all possible objects that may be found in the collision region, controlling the activation of the RTOD application, and efficiently stopping the detector motion.

Text communication is often perceived as lacking some components of communication that are essential in sustaining interaction or conversation. This interaction incoherency tends to make 
text communication plastic. It is traditionally devoid of intonation, pitch, gesture, facial expression and visual or auditory cues. Nevertheless, Instant Messaging (IM), a form of text communication is on the upward uptake both on PCs and on mobile handhelds. There is a need to rubberise this plastic text messaging to improve co-presence for text communications thereby improving 
synchronous textual discussion, especially on handheld devices. One element of interaction is gesture, seen as a natural way of conversing. Attaining some level of interaction naturalism 
requires improving synchronous communication spontaneity, partly achieved by enhancing input mechanisms. To enhance input mechanisms for interactive text-based chat on mobile devices, 
there is a need to facilitate gesture input. Enhancement is achievable in a number of ways, such as input mechanism redesigning and input offering adaptation. This thesis explores affective gesture mode on interface redesign as an input offering adaptation. This is done without a major physical reconstruction of handheld devices. This thesis presents a text only IM system built on 
Session Initiation Protocol (SIP) and SIP for Instant Messaging and Presence Leveraging Extensions (SIMPLE). It was developed with a novel user-defined hotkey implemented as a one-click context menu to &ldquo
fast-track&rdquo
text-gestures and emoticons. A hybrid quantitative and qualitative approach was taken to enable data triangulation. Results from experimental trials show that an 
Affective Gesture (AG)approach improved IM chat spontaneity/response. Feedback from the user trials affirms that AG hotkey improves chat responsiveness, thus enhancing chat spontaneity.